The politics of streaming time: When elastic time meets troubled tiers in the turn to the subscription video-on-demand (SVOD) model

Abstract

This article proposes the politics of streaming time as a productive analytical lens to unflatten more general accounts of nonlinearity and acceleration, to reveal the mediating role of streaming services and broader power relations around terms of access in value reconfiguration, and to critique the cultural-economic logics of the streaming economy. Building on the literature on media and time in the internet society, and integrating platform studies and television studies, this article analyses the politics of streaming time with reference to Chinese streaming services’ turn to the subscription video-on-demand (SVOD) model, a process negotiated among multiple parties including streaming services, advertisers, audiences and regulators. The analysis shows how streaming services orchestrate elastic schedules in concert with flexible pricing and dynamic bundling to manipulate temporal and affective experiences of viewers while causing contingent boundaries of subscriber benefits. It explains how the data-contingent approach confronts subscribers with a boomerang moment of temporal displacement and surge pricing. Further, it reveals how multiple temporalities including artificial ephemerality, collusive synchronicity and metric thresholding as well as reconfigured televisual linearity are mediated to entice and extract value from subscribers. The conclusion reflects on how exploring the politics of streaming time allows a more textured understanding of emerging temporalities and the negotiation around terms of access in an increasingly contingent cultural market.

Introduction

Time is fundamental to value production on video streaming services. As captured by terms such as on-demand and time-shifting, nonlinearity as a major value proposition is often associated with the notion of audience agency, the freedom to navigate a vast catalogue of content offerings (Johnson, 2019; Lotz, 2017; Tryon, 2015). Netflix, with its signature binge-release strategy, pioneers in prioritising self-scheduled viewing and becomes synonymous with the binge-viewing culture associated with a sense of choice and immediacy (Jenner, 2018; Pilipets, 2019; Turner, 2019). The embrace of binge release does not, however, spell a decisive departure from the legacy practice. Streaming services, including Netflix more recently, have experimented with paced scheduling from weekly to periodic release. Temporality, cultivated in various ways, continue to raise questions about media distribution and value production in the realm of streaming services. While existing research has highlighted the affordances of nonlinearity and binge-viewing, how multiple temporalities are produced by and productive of streaming services remains less known.

Adding complexity to shifting temporal regimes is the highly diversified and contingent nature of streaming services, as their business models, content strategies, and distribution practices continue to evolve. Categorical distinctions dissolve between advertising-supported, subscription-funded and transaction-based services. Different practices and politics of content aggregation work to reshape the value of content (Vonderau, 2015). Pricing structures change, and different pricing strategies often co-exist in an ongoing process of experimentation, contestation and normalisation. From ‘free’, freemium and subscription to personalised pricing, pricing models in digital media vary considerably with different ways of distribution (Lobato, 2021). As streaming service providers exercise power over what, how, at what price and pace content is made available, they have profound implications for how audiences experience and value culture.

In this rapidly evolving space, what kind of temporalities are mediated and promoted by streaming services? How do temporal dynamics co-evolve with various aspects of distribution practices such as user engagement and pricing approaches? In what ways do these practices reconfigure the value regime and industrial logic? And what do these mean for audiences?

This article seeks to probe these questions by approaching the politics of streaming time as a key analytical entry point. The politics of streaming time, as this article will show, provides a productive lens through which to examine the distribution of power and control in the streaming economy. A critical analysis of the politics of streaming time invites us to consider actually existing, multiple temporalities and their cultural-economic logic mediated by streaming services and negotiated among multiple stakeholders despite the predominant discourse about ‘on-demand’ culture. It provides an opportunity to revisit the crucial role of time and temporalities in the business and culture of television as the industry boundaries continue to shift. It also helps to reveal how multiple temporalities intersect with other distribution approaches such as pricing, bundling as well as user engagement in value reconfiguration.

This article focuses on the specific context when the once largely advertising-supported streaming services turn to the subscription video-on-demand (SVOD) model with their own take in China. The case of China is useful not only for the surging popularity of video streaming services in the market, but more importantly, for the opportunities it affords to conceptualise multiple temporalities at play and their change over time. The continuing experimentation in distribution practices in the market offers a fertile ground to explore the negotiation around the shifting temporal logic among multiple parties and the reconfiguration of the value regime. Moreover, China offers a helpful case to expand our knowledge about the varying trajectories of streaming services across the world and their underlying logics. The discussion here complements the existing literature on SVOD services by mapping the trajectory of the subscription model as it is introduced in the China market. It seeks to contribute to the broader literature in accounting for ‘rates of change’ (Cunningham and Silver, 2013) in the relationship between temporality and streaming media, and digital transformation of media distribution more broadly (e.g. Bruun, 2019; Doyle, 2016; Johnson, 2019; Lobato, 2019; Lotz, 2017; McDonald et al., 2021; Perren, 2013). To develop the analysis, this article draws on publicly available data reflecting scheduling strategies and pricing approaches of leading video streaming services in China since the introduction of temporally anchored subscription benefits and transaction tiers in the broader context of their pivot to the subscription model. The examples here are chosen for their high-profile status in the process of experimentation, including both successes and failures. This is complemented by media articles, industry reports, as well as user reactions on social media. Relevant material is selected for thematic focus and temporal scope, specifically on the transition to the SVOD model, the shifting approaches to scheduling and pricing by leading streaming services including iQiyi, Tencent Video and Youku Tudou between 2015 and 2022, a timeframe chosen to capture the spectrum of approaches experimented by these services.

The rest of the article begins with a brief overview of the literature on time and media in the context of internet society and streaming culture. Then it contextualises the turn to the subscription model by the once predominantly advertising-supported streaming services in China, explaining what drives this shift and how temporal acceleration underpins streaming services’ attempt to maximise subscriber revenue. Next, it discusses how the design of temporal acceleration is increasingly implicated in hybrid pricing schemes and dynamic content bundling, causing contingent boundaries of subscriber benefits while manipulating temporal and affective experiences of viewers. It explains how the data-contingent approach paradoxically constitutes a boomerang moment of temporal displacement and surge pricing for subscribers. Further, it reveals how multiple temporalities including artificial ephemerality, collusive synchronicity and metric thresholding are mediated to entice and extract value from subscribers. Finally, it critiques the failed attempt of reimposing televisual linearity in the super-early access bundle, which intends to maximise economic value capture yet compromises the cultural norm.

Time, temporality and video streaming

Time, a fundamental concern in society at large, has attracted growing interest in media and communication research over the past decade (Keightley and Punathambekar, 2021). The conceptualisation of time during the early years of the internet evokes a universal notion of time. For Castells (2000), network society has ushered in an age of ‘timeless time’, breaking down the biological sense of time and the logical sequence of time. Going beyond the framing of an atemporal or homogenous temporal order, researchers have drawn attention to a multiplicity of temporality in the internet age that calls for more research (Leong et al., 2009). Despite a predominant focus on the tempo of acceleration, as Wajcman (2015: 9) contends, digital media ‘make possible new and multiple temporalities’. Moreover, researchers have done important work at the intersection of temporality, technology and power. In Sharma’s (2014: 9) words, multiple temporalities exist concurrently ‘in a grid of power relations’. More recently, a growing body of research has advanced a more nuanced debate on mediated time by teasing out how different technologies, devices and platforms structure temporalities in various ways. For example, Weltevrede et al. (2014) provide an insightful analysis of how social media platforms fabricate distinctive real-time temporalities. As they argue, ‘the pace within each engine and platform is internally different and multiple in itself’ (ibid: 135). Coleman (2018) complicates our understanding of the present by explaining the production of diverse present temporalities on social media. In the case of Netflix binge watching, for example, ‘the progression from past to present to future is suspended, and nextness or pre-emergence becomes absorbed within a kind of stretched or expanded present’ (ibid: 613). Focusing on algorithmic media, Bucher (2020) convincingly argues that its temporal nature lies in the notion of right-time, operating as a function of relevance. This body of literature provides rich insights and invites further investigation into the production and politics of multiple temporal dynamics in contemporary media.

As video streaming services become key mediators of culture, they open up opportunities to pursue this line of inquiry further. Meanwhile, it is helpful to situate the reconfiguration of the value regime around temporal dynamics in continued exercises over time. In broadcast television, the centrality of time is evident in both serial narrative construction and scheduling optimisation (Ellis, 1982; Williams, 1974). Scheduling is closely tied to the production of audience commodities and the advertising-supported business model in commercial television (Ang, 1991). Later, technologies such as VCR and DVD appeal to the audience with a promise of overcoming ephemerality of television (Cubitt, 1991; Kompare, 2006), where viewers can obtain box sets through rent or purchase. The advent of internet distribution technologies enables users to download and share content with a click. As viewers increasingly stream content online, the notion of ownership gives way to the culture of access, via advertising-supported, subscription or transaction-based models. While binge release has undoubtedly disrupted the linear consumption pattern typically associated with broadcast television (Jenner, 2018; Johnson, 2019; Lobato, 2019; Lotz, 2017), linear logics continue to structure streaming services (Johnson, 2019) and legacy practices such as staggered release persist in distribution strategies (Sanson and Steirer, 2019). Experiments continue as a growing range of streaming services vie for audience share, and equally importantly, time share. This body of literature provides valuable groundwork for further research on how streaming services remediate earlier forms of screen media technologies in temporal configuration.

While these works have provided rich insights, the multiple temporalities produced by and productive of streaming services, the underlying logic and power relations in reconfiguring the value regime remain understudied. To address this gap, this article draws attention to the politics of streaming time. It does so by building on the lines of inquiry into media time and temporalities in the internet society, and integrating literature on television studies and platform studies. It proposes a conceptual approach to streaming time by moving beyond the more general accounts of on-demand and nonlinear temporality, and complicating the binaries of linear and nonlinear. This helps to nuance our understanding of the emerging temporalities in the streaming economy. Moreover, it argues for the need to situate the temporal dynamics in ‘the politics of platforms’ (Gillespie, 2010) to account for how mediated temporalities are connected to platform power. Equally importantly, considering video streaming services in relation to earlier forms of screen media technologies and practices helps to critically assess both change and continuity in the temporal logics.

The next section turns to the pivoting moment as Chinese streaming services attempt to attract subscribers, which affords the opportunity to examine how, why, and with what consequences multiple temporalities are mediated in an evolving industry context, and how emerging temporalities are entangled with power dynamics.

The pivot to the subscription video-on-demand in China

Audience commodities in flux

For a long time, video streaming services in China were predominantly supported by advertising where audiences were offered free services in exchange for their attention and data. Considering the broader societal norm, China was a market known for low willingness to pay for internet content among users and prevalent informal circulation of content (Karaganis, 2011). The reliance on advertising met a turning point when leading video streaming services in China started to introduce the subscription model. iQiyi began the experiment to seek subscription revenue in 2011, a move followed closely by Tencent Video and Youku Tudou. In 2022, China became the second country, trailing only the United States, to cross the threshold of $10 billion subscription revenue from video streaming services (Statista, 2022). The market has evolved to a stage where both subscription and advertising constitute major revenue sources.

The introduction of the subscription model is a response to both the growing content cost and the stagnating advertising revenue. First, the growing cost in content acquisition and production. Following the formalisation of the once piracy-ridden market, intensive competition in licensing deals has pushed up content acquisition cost significantly (Zhao and Keane, 2013). Moreover, streaming services have continued to invest in content production under various original programming labels to reduce reliance on licensed third-party content, expand offering, and develop their competitive edge (Keane, 2015; Zhao, 2019).

Second, advertising as the main revenue stream has faced intense competition from short video streaming services such as Douyin and Kuaishou offering algorithmically distributed user-generated content. The growing user base and the integration with e-commerce incentivise advertisers to explore short video streaming platforms for potentially more effective advertising, typically tasked to boost sales figures. When compared with display advertising mainly for branding purposes, a higher level of integration with e-commerce means a shorter path to driving purchase behaviour. In such context, advertisers have reduced budget allocation to long-form video streaming services while increasing spending on short video platforms (The Paper, 2022). For streaming services like iQiyi, Tencent Video and Youku Tudou, this shift in advertisers’ budget allocation becomes more concerning when advertising as a dominant revenue stream has yet to deliver convincing profitability.

In this context, converting ad-supported viewers into paying subscribers becomes the answer. For service providers, the subscription model is an option to enable a direct relationship with customers and provide predictable up-front revenue streams in unpredictable contexts such as the cultural industries (Picard, 2011). As Arditi points out, digital subscriptions constitute a quintessential part of ‘unending consumption’, a form of ‘the expansion of the means of consumption’, where ‘an individual’s consumption becomes consistent and constant’ (2021: 308). To appeal to viewers, SVOD services draw on customisation approaches while creating an appearance of limitlessness (Lotz, 2017; Rodríguez Ortega, 2023).

Temporal acceleration as subscriber benefits

To fully appreciate value formation around the pivot to the subscription model, however, requires us to look beyond content appeal. In fact, the evolving distribution model and the temporal dynamics are key to understanding the turn to subscribers in China. Temporal acceleration is typically offered as subscriber benefits to drive subscriber growth. Under an accelerated schedule, subscribers gain earlier access to newly released content. This can take the form of access to more episodes at regular intervals or the possibility of bingeing through to the end.

A high-profile experiment at an early stage was iQiyi’s The Lost Tomb (Daomu Biji) (2015), a web series with the iQiyi original label adapted from a highly popular fantasy adventure internet novel. At a time when the streaming service was still testing the validity of the subscription model, the existing fan base and the well-established cultural norm of subscription in internet literature consumption presented an opportunity. To drive subscriber growth, distribution ran on two tracks. While non-subscribers can only view one episode per week, subscribers had the privilege to speed up and binge watch the full series in the fourth week. The series proved to be an immediate hit, with fans rushing to sign up as subscribers for the privilege of accessing the full series earlier (Global Times, 2015). This is regarded as an encouraging signal of Chinese users’ willingness to pay for video streaming services, and relatedly, a strong potential in the SVOD model. Clearly, temporal acceleration in access to content plays a crucial role in driving subscriber conversion. Since then, streaming services have increasingly adopted the tiered distribution model, offering early-access or binge-viewing privilege to subscribers.

In the context of hybrid business models and multi-site distribution, streaming services draw on temporal acceleration to appeal to potential subscribers while maintaining accessibility for ad-supported audiences without disrupting the pricing norm among the latter. The rhythmic variance in distribution, associated with different pricing models and viewing behaviour, speaks to the perceived value of content for various audiences. Temporal acceleration capitalises on fans’ enthusiasm and the sentiment of fear of missing out or lagging behind. The aesthetic of immediacy allows viewers to connect to an unfolding story as it happens. This speaks to the broader picture where fans assume a crucial place not only in production decisions (Fung, 2019) but also in distribution approaches (Zhao, 2021). In the pivot to the subscription model, fans are targeted as potential subscribers for their desire for the temporal lead in content consumption as a form of cultural capital. Fans’ wish to stay in sync with the latest content is well-documented. This has been well served, for example, by fansubbing groups in China who act as efficient cultural intermediaries in the video consumption culture (Hu, 2012).

The temporal tiers further cement the power of streaming service providers in the broader screen ecology. As investment in original content production sees streaming services increasingly licensing content to broadcasters, this lends the former more negotiation power to maintain the lead in temporal windows for subscribers. When content is available on both streaming services and broadcast television, temporal windows are maintained so that streaming service subscribers can stay ahead of non-subscribers and broadcast audience. This serves the streamers’ purpose of building distinction from broadcasters not only through original content, but also through distribution approaches and associated temporal privileges for subscribers.

Dynamic unbundling and the boomerang moment of temporal displacement and surge pricing

Building on the success of offering early access to drive subscription, streaming services further launched the subscriber-facing super-early access feature. Rather than a separate subscription tier, this feature is only available through transaction on top of existing subscription packages, typically introduced for the final few episodes of highly popular titles. Streaming services usually provide the option to unlock more episodes in an accelerated schedule or binge through to the end. This experiment began at a time when subscriber growth started to stagnate around 2019. In this context, driving revenue growth from the existing subscriber base became a priority. In 2020, 30% of web series and 59% internet-first drama series adopted this model, with 30% video streaming users spending between 10 and 50 yuan for super-early access (China Netcasting Services Association, 2021).

When the industrial logic of upgrade culture meets the culture of accelerated consumption cultivated since the turn to the subscription model, subscribers are trapped in a dilemma. They either must pay a premium to upgrade or be excluded from continuing acceleration. Tencent Video experimented this model with The Untamed (Chen Qing Ling) in 2019, pricing the final six episodes at 30 yuan in total or 6 yuan per episode on top of the subscription package. The experiment with the super-early access feature strategically targets this web series adapted from a highly popular internet danmei novel. Featuring stories of romantic relationships between men, the genre of danmei, also known as Boy’s Love, has developed an enthusiastic fan base who appreciate alternative masculinities and sexualities in the heteronormative Chinese society (Zou, 2022). Despite the controversial decision which initially saw some subscribers calling for boycotting on social media against the surprise paywall, over 2.6 million viewers paid for early access to the finale on the first day it was made available (Wen and Bi, 2019). As a result, Tencent Video raked in over 78 million yuan on top of the subscription revenue over one night. This attests to the strong allure of temporal lead and the coercive power of upgrade culture. Since then, the on-demand super-early release feature has become more widely adopted. For streaming service providers, the potential to extract further revenue from existing subscribers extends a lifeline in the context of stagnating subscriber growth.

Temporal elasticity in the super-early access feature is closely associated with the logic of dynamic unbundling. The subscription model is based on the all you can eat-logic, which attaches value to the bundle offered by the streaming service (Colbjørnsen et al., 2022). Adding an additional layer of transaction-based access on top of the subscription package essentially constitutes unbundling, where the value of an episode is detached from a series. As streaming services configure how bundling or unbundling works, and equally importantly at what pace and cost, the value anticipated in the subscription package may dissipate for subscribers as streaming services attempt to augment value capture.

In the restructured tiers of access, differentiated paces may prompt a sense of temporal displacement for existing subscribers. Facing the bifurcation of the release schedule, subscribers have to recalibrate their sense of time. As Sharma notes, recalibration involves various ways of adapting ‘senses of the future or the present, to an exterior relation’ (2014: 18). For subscribers who do not upgrade to the transactional tier, temporal displacement results from a manufactured lag, where they experience a perceived deceleration when being relegated to a lower tier in the temporal hierarchy. Although the pace of schedule remains the same for them, recalibration entails not only subscriber-content temporal relations, but also the social experience of content consumption, such as the way they engage in cultural conversations in the viewing community including the need to avoid spoilers. For those who pay for the transactional tier to stay current, they recalibrate to the temporally compressed mode of viewing. This places these subscribers in ‘a kind of stretched or expanded present’ (Coleman, 2018: 613). The reorientation around an accelerated pace of release may further involve possible adjustment of rhythms in other aspects of their everyday life, from biological to interpersonal and social, so that they can stay in sync with the latest content at a premium cost in time.

While regular rhythms on streaming services allow for predictability, the transactional tier is designed to be contingent. Specifically, whether this tier will be introduced, when this will happen, and at what cost. The contingency and ambiguity embedded in terms of service lend streaming services convenience in modulating content access over time at will in their own interest. Although early access is highlighted as a subscriber privilege, there is no transparent nor consistent standard. In fact, while initial experiments typically apply to the final few episodes, streaming services increasingly utilise moments of cliffhangers or narrative climaxes to introduce the super-early access even earlier. Take the smash hit Joy of Life (Qing Yunian) (2019), for example. The transactional tier for super-early release kicked in around halfway through the 46-episode series, when subscribers can unlock six more episodes per day for an additional 50 yuan. To put this in perspective, monthly subscriptions cost 19 yuan on iQiyi and 15 yuan on Tencent Video at the time. The uncertainty arising from dynamic unbundling impacts on user experience regarding the terms of access to content, both temporally and pricewise.

What underlies temporal uncertainty in dynamic unbundling is the data-contingent approach in launching the super-early access tier, typically deployed for highly popular content. Indeed, the super-early access tier has become a new status marker for content on streaming services. For video subscription platforms, evaluating the value contribution of each content piece is essential to managing content portfolios (Kübler et al., 2021). Various data analytics inform the evaluation of potential content popularity and relatedly the prospect of revenue capture from subscribers. First, as internet literature-to-streaming adaptation becomes a prominent content approach to expand subscriber base, various readership and engagement data on internet literature sites informs not only content acquisition and production decisions but also distribution strategies for web series. Streaming services tend to target hit titles on internet literature sites and introduce the super-early release tier for the web series.

Moreover, the decision to introduce additional transactional tiers for subscribers is contingent on real-time audience reception as reflected by popularity metrics. In fact, major streaming services have become a key part of audience measurement infrastructure with their own popularity indexes informed by various viewership and engagement metrics. This can be seen as an attempt to solidify platform power in the streaming ecology, which draws on both economic and cultural logic to mobilise datafied affective labour among audiences including fans in particular (Zhao, 2021). For example, iQiyi’s in-house content popularity index plays a key role in scheduling decisions as super-early access becomes a common on-demand feature for popular titles. In the case of My Heroic Husband (Zhui Xu) (2021) streamed on iQiyi, the service provider decided to introduce the super-early release feature when its popularity index hit 10000, a benchmark for hit titles on iQiyi. Therefore, data is not only symbolic but also operational, feeding into systems of audience measurement to shape the terms of access to content. The centrality of datafied fandom activities cultivated by streaming services is evident in shaping the temporal regime of viewing culture and value extraction.

Considering various forms of datafied labour feeding into metrics and analytics of content popularity, the super-early access tier essentially unleashes a boomerang effect for subscribers. Subscribers are confronted with not only a dilution of temporal lead, but also an introduction of temporal uncertainty. This complicates what Rodríguez Ortega (2023) points out in his analysis of Netflix, that subscribers pay for themselves in the context of the algorithmic process. In the case of data-contingent super-early access feature, the issue at stake is not content recommendation, but the terms of access to content, including when and at what cost. As Bucher (2020) argues, algorithmic media and feedback loops usher in a distinctive right-time temporality, where platforms dispense the promise of delivering content at the right-time. This is underpinned by both socio-technical construction and user modulation of right-time (ibid: 1704–1706). What constitutes right time can indeed be contested. This becomes more complicated when dynamic unbundling works in connection with flexible pricing schemes. While personalisation in streaming services is more often discussed in relation to content recommendation (e.g. Echauri, 2023; Lobato, 2019; Lotz, 2017; Webster, 2023), the temporal dimension of content distribution and the pricing structure complicates the customisation process which shapes the terms of content access.

Indeed, the boomerang moment is already written in the script as audiences take part in constructing temporalities on streaming services. Datafied labour among users becomes a part of the feedback loop, which in turn informs how tiers of access are to be configured in terms of temporality and pricing. As Athique insightfully points out, data production by audiences becomes enshrined into transactions, a process driven by the pursuit of ‘an integrated commodity’ (2020: 560–561). In this process, ‘unwitting labour’ by users makes data inputs and outputs fictious commodities (Athique, 2020: 561). In the case analysed here, the unwitting labour is sold back to targeted users. Through dynamic unbundling, the transactional tier on top of the subscription package creates fictitious commodities out of data production by viewers and fabricates a desirable temporality tied to priority access to content. Beer (2022) vividly conceptualises the repeated process of data analytics and multiplication of looping as data coiling, characteristic of a recursive society. Within this context, as he argues, ‘outcomes are already implicated by the looping of prior decisions and the data that those decisions have recursively produced’ (ibid: 4). In the case of reconfigured terms of access on streaming services, datafied audience labour in turn validates and propels accelerated consumption, which informs future manoeuvres in temporal and pricing strategies as a part of service to be experienced by users.

Multiple temporalities in service of temporal acceleration: Artificial ephemerality, collusive synchronicity and metric thresholding

While temporal acceleration becomes central to value articulation in subscription and transaction-based offerings, the reconfigured temporal dynamics are more complicated than it seems. Underpinning temporal acceleration are multiple temporalities constructed by streaming services in the scheduling practice. From artificial ephemerality to collusive synchronicity and metric thresholding, these serve the interest of streaming service providers as they seek to augment value capture from subscribers.

First, streaming services create artificial ephemerality that works in connection with the reconfigured temporal windows to nudge users into the super-early access tier. A typical approach is to dispense limited-time offerings of promotional deals, often with bundling techniques. Ephemerality creates a sense of urgency, prompting an increased likelihood of user buy-in. In the case of Joy of Life, the promotional package for super-early release was on offer for 1 week. Upon its expiration, existing subscribers wishing to watch the rest of the episodes in advance had only one option of paying 3 yuan per episode for all the rest 25 episodes. This amounted to 75 yuan for existing subscribers. In comparison, the package deal was presented as offering more value. Indeed, internet television has not ceased to be an ephemeral medium, especially in terms of promotional strategies. As Grainge notes, ‘television is now less ephemeral in the evanescence of programme content but much more ephemeral in the brevity of the promotional and paratextual forms that surround, mobilise and give meaning to that content’ (2011: 7). As it encourages viewers to seize the moment, artificial ephemerality further cultivates the culture of binge-viewing. Habituating users to binge-viewing increases user dependency on platforms, which lends more power to platforms as they reconfigure content distribution in their own interest.

Moreover, different streaming services orchestrate temporal synchronicity, which leaves subscribers no alternative when facing the data-contingent approach to content release and surge pricing. Given a relatively high level of market concentration, streaming services tend to act concertedly in introducing the super-early access feature when a series streams on more than one service. Collusive synchronicity leaves subscribers with no alternative but to accept either the price hike or a compromised temporal lead. For example, when Joy of Life was simultaneously streamed on both iQiyi and Tencent Video, the two service providers not only agreed on the tiered schedule but also concertedly introduced the super-early access tier with the same terms. Despite complaints by some subscribers, others have unlocked the tier, willingly or less so. Essentially, prices carry meanings. The discrepancies in the perceived value result in divergent opinions in terms of what a fair price is, which makes surge pricing possible in the first place. Through collusive synchronicity, streaming services work at the market level to manipulate temporal experiences of subscribers in their own interest.

Further, streaming service providers employ metric thresholding to drive audience engagement, where an accelerated schedule is launched only when the threshold is met. Temporal thresholding operates as a rhythmic device through the key metric of viewing hours. As Seaver (2019) insightfully points out, the use of captivation metrics such as time spent viewing entrap audiences as streaming services aim to maintain users’ attention. Here, the temporal metric captivates users further when it is tied to the prospect of an accelerated schedule to prompt user engagement in particular ways. For instance, when Word of Honour (Shanhe Ling) (2021) became an instant hit on Youku Tudou in the first week of release, fans sprang up on social media to call for an accelerated pace of distribution. Instead of responding to audience demand with an updated schedule straight away, Youku Tudou set up a temporal threshold for viewers to collectively cross before an accelerated schedule can be released. The threshold was set up in connection to the metric of total viewing time, and relatedly, subscriber numbers. Viewers were encouraged to recommend the series to friends to drive subscription and hit viewing time targets so that they can unlock an accelerated schedule. The networked race towards the temporal threshold eventually prompted a faster pace of content release for subscribers, with two more episodes released weekly over the weekend. Meanwhile, the streaming service introduced the super-early release feature to capture further value.

As scheduling is punctuated with metric thresholding, it operates as a rhythmic device, in terms of both content release and audience engagement. More importantly, metric thresholding establishes a connection between the pace of schedule and the intensity of audience engagement, as well as between personal rhythm and networked rhythm. Through the introduction of a suspension, temporal thresholding orients viewers around the possible future of acceleration. It instils a temporal aspiration which motivates viewers to engage with content and other viewers in particular ways, from signing up as subscribers to making recommendations to friends to hit temporal targets. This allows streaming services to drive subscription and engagement, and equally importantly, to gather data which feeds further into decision making. Moreover, it evokes a sense of liveness and a ‘common public time’ usually associated with broadcast television (Scannell, 2014: 356). Indeed, despite the advent of time-shifting technologies, the value of liveness persists (Levine, 2008). As Van Es (2016) posits, liveness takes on new forms as it is mediated as a result of a dynamic interaction between media institutions, technologies and users. As viewers are invited to contribute to the progression of a live event with its impact on scheduling, the streaming service creates a perceived sense of viewer agency. Equally importantly, it lends viewers a sense of co-presence and community as they work towards the threshold. The viewing community is connected not only through content, but also via networked efforts in making content accessible at a desirable pace. It is the shared sense of time in these dimensions that lends a special significance to the moment of release when the viewing community crosses the temporal threshold. As a rhythmic device to engage viewers around temporal negotiation, metric thresholding orchestrates a networked rhythm of viewing and datafied sociality. In what Lefebvre calls a ‘polyrhythmia’ (2004: 89) where multiple rhythms take place at different paces come together, networked rhythm not only parallels personal rhythm of viewing, but the two is enfolded into each other and interact with the pace of scheduling.

As artificial ephemerality, collusive synchronicity and metric thresholding introduce variations to regular and predictable rhythms, audiences are oriented around potential changes. These multiple temporalities can be seen as designed pre-emergence, ‘the just-beforeness that is nevertheless an integral aspect of the present’, where affects and feelings are mobilised (Coleman, 2018: 167). As temporalities are constituted on and across platforms, they evoke and mobilise audience sentiments ranging from anticipation and hope to excitement or frustration, and prompt specific viewer engagement. Audience expectation and experience of temporal malleability is enmeshed with fluidity between personal time and network time, between leisure and labour, and simultaneously between agency and constraint. Multiple temporalities constitute a key aspect of the new commodities in the subscription streaming economy where ephemerality, synchronicity, suspension and acceleration are variously brought into value production.

The contested televisual linearity in the super-early access bundle

As streaming services evoke multiple temporalities in service of temporal acceleration, frictions can arise at their intersection. This section turns to the controversy around the coercive attempt to re-impose linearity in the super-early access bundle. Specifically, the episodes are de-bundled from the series first, before being sold as a separate bundle for super-early release. For subscribers hoping to unlock content behind the super-early access paywall, they must pay for all the rest episodes instead of only paying for certain episodes of their choice such as the finale. Under this re-bundling approach and pricing structure, users cannot skip ahead without paying for the whole package.

Contrary to the common association with time-shifting experiences, the episode-level re-bundling in the transactional tier on top of the subscription package confines subscribers to televisual linearity. For example, in the case of the 28-episode crime drama Crime Crackdown (Saohei Fengbao) (2021), Tencent Video introduced the super early-access feature from episode 15, and existing subscribers must unlock the rest of the series episode by episode at 3 yuan each. While priced per episode, the super-early access was essentially sold as a bundle. In other words, this prevented users from accessing any episode without paying for the package.

This approach has its discontents. Many subscribers vented their complaints on social media, considering it as unfair, exploitative, and violating the freedom of choice. Indeed, the idea of a fair price exists in consumer perception that ‘the price contains a sound judgment about the economic system and the moral glue that keep it together’ (Sanz, 2014: 134). While Tencent Video anticipated the demand for temporal acceleration, the pricing model misaligned with user expectation regarding the temporal protocol as well as the pricing norm. Deprived of the freedom to navigate content in the sequence that they like while facing a price premium, subscribers can only submit themselves to what they see as an unfair price structure or the patient pacing of a regular schedule as dictated by the service provider.

The violation of user expectation becomes more problematic considering the long-established culture of achronological viewing. As Kompare (2006) suggests, television is increasingly accessed via files, with a key temporal difference from the flow logic of the broadcast form. For Kelly (2011: 125), streaming is the application of the logics of flow to a medium of files. This foregrounds the mediating role of streaming service providers in organising flows. To recast linearity is to reintroduce rigidity to a viewing culture where user agency is closely associated with temporal fluidity. Adding complexity to this is the pricing scheme, where tiering operates in connection with the logic of scheduling and bundling. Price has its symbolic value (Sanz, 2014), and when misaligned with the cultural logic of market, leads to the perception of illegitimacy. Here, the industrial logic of recasting linearity together with bundled pricing violates the cultural logic of viewing practices on streaming services. Further, considering the short history of subscriber conversion on streaming services in a market long dominated by free, advertising-supported viewing in China, lower price points for regular release or content circulating in the informal market further raise the question of value regarding temporally differentiated pricing. As Lobato notes, consumer attitudes towards pricing models of subscription video-on-demand services depends very much on ‘[the] historical norm about what television is and how it should be consumed and financed’ (2021: 322).

In addition to drawing user pushback, the super-early access feature attracted a drumbeat of criticisms from both state media and consumer protection agencies at local and national levels. People’s Daily, for example, denounced the practice and called for more regulation (People.cn, 2019). Shanghai Consumer Council criticised the bundling approach to super-early access for neglecting consumers’ right of choice (The Paper, 2021). This further triggered a commentary from China Consumers Association, calling for industry self-reflection and rectification (Sina.cn, 2021). Under such pressure, leading streaming service providers including iQiyi, Tencent Video and Youku Tudou concertedly announced the decision in October 2021 to discontinue the super-early access feature. While media and regulatory oversight have their immediate effect, the long-term impact remains circumscribed. Similar practices of manipulating temporal dynamics in connection with surge pricing have emerged in various disguises. In addition to on-demand finale sold on its own, ticketed watch parties with main cast through livestreaming, for example, allow subscribers to access final episodes in advance. Negotiations continue to play out between subscribers and streaming services at the intersection of the temporal order, the price structure and the bundling logic.

Conclusion

As video streaming services continue to evolve in the context of multi-tier distribution and hybrid business models, they shape the terms of access to content, temporally, economically, and culturally. Existing literature has identified various ways in which streaming services wield distribution power, from propelling consumers into time-shifted and binge-viewing experiences to customising content recommendation to pique their interest. Yet, a more nuanced understanding towards the construction of temporalities, their co-evolution with distribution practices in shaping the value regimes, as well as their impact on viewers remains lacking.

To address these gaps, this article draws attention to the politics of streaming time as a productive analytical lens. In doing so, it makes several contributions. First, by examining how streaming services mediate multiple temporalities, the article demonstrates the need to move beyond the more general accounts of on-demand and nonlinear temporality, often with a binary approach. It also reveals how multiple temporalities intersect with other distribution approaches such as pricing, bundling and user engagement in value reconfiguration. Second, by adding a temporal dimension to ‘the politics of platforms’ (Gillespie, 2010), it demonstrates how the politics of streaming time is integral to understanding not only the mediating role of streaming services but also the broader power relations in the streaming economy. Third, the discussion contributes to a growing body of literature which offer critical assessment of changes and continuities in media distribution, particularly since the advent of streaming services. Finally, the empirical analysis complements the existing literature on subscription video-on-demand services by further the understanding of the trajectory and impact of the subscription model in the China market. While empirically grounded in the industry in China, the study has broader relevance by demonstrating the need to critically investigate the politics of streaming time in value reconfiguration in other markets to assess broader patterns.

As Chinese streaming services turn to subscribers in a market long dominated by the advertising-supported model, it constitutes a re-commodification process in response to the growing content spending and the declining advertising revenue. Beyond content appeal, distribution is a key site of value production where temporal dynamics are reconfigured in connection with pricing models and content bundling approaches. The culture of temporal acceleration and the aesthetics of efficiency appeal to the sentiment of anticipation, enthusiasm and fear of missing out among viewers to drive subscriber growth. As viewers are enticed to sign up as subscribers, streaming services further habituate these viewers to a culture of temporal compression while cementing their competitive advantage to broadcasters.

The terms of acceleration, however, can give rise to troubled tiers in hybrid business models. Promoted as an on-demand feature for subscribers, the transactional super-early access tier on top of the subscription package operates at the intersection of temporal acceleration and upgrade culture. Its deployment is underpinned by dynamic unbundling and contingent on data-inflected content popularity. While denoting an aura of legitimacy tied to the social value of content and the carefully cultivated culture of accelerated consumption, it confronts subscribers with temporal displacement and surge pricing. This paradoxically constitutes a boomerang moment for subscribers as a result of the recursive, data-contingent logic in temporal reconfiguration. The dilemma becomes more troubling considering subscribers have already acclimatised to accelerated consumption.

As streaming services cultivate a culture of temporal acceleration in their own interest, they in fact usher in multiple temporalities. Artificial ephemerality, collusive synchronicity, and metric thresholding are enmeshed with, and more importantly, steers a culture towards temporal acceleration in a hybrid scheme of tiered access to content. These temporal mechanisms operate at both micro and macro levels, in connection with platform-specific and cross-platform practices. They work across personal and network time to manipulate temporal and affective experiences of viewers. Together they constitute coercive temporal structures to serve the interest of streaming service providers.

While televisual ephemerality and liveness demonstrate continuing relevance, friction can arise at the interface between televisual linearity and temporal acceleration. An illustrative example is the attempt to recast televisual linearity in the super-early access bundle. Although streaming culture is largely associated with nonlinearity, this should not blind us to the various ways streaming services repackage linearity and sell it back to audiences. The super-early access deal, when introduced with a rigid temporal linearity and sold as a bundle, embraces the chronological authority typically associated with broadcast television. It violates the common expectation of streaming services in relation to viewing autonomy and a long-established norm outside the purview of broadcast television. When the industrial logic of temporality is divorced from the cultural logic of viewing practice, the pricing scheme is deemed as manipulative and illegitimate.

Rather than a linear process, the co-evolution of temporalities and pricing models as well as content bundling approaches has witnessed multiple reconfigurations as Chinese streaming services turn to subscribers. While various pricing schemes seem to offer options, temporal regimes on streaming services are intentionally curated, which can be less premised on individual consumer choice and viewing autonomy than what platforms claim. The experiments are marked by a sense of paternalism, steering consumption culture in particular ways more in streaming services’ interest. As distribution models become more hybrid, temporal malleability operates in connection with price elasticity and bundling flexibility to capitalise on both the desire for and the habituation to accelerated consumption. As streaming services continue to evolve, how the politics of streaming time unfold in the access economy in similar or different ways in various contexts warrants further research. Future research might also explore further how this reconfigures viewing experience including the specific ways viewers come to terms with multiple and contingent temporalities.